Nicole de Silva

Establishing norms for intestinal length in children.

BACKGROUND Existing data on pediatric intestinal length (IL) are limited because most studies report postmortem values. Using prospective data, appropriate norms for IL were developed. METHOD The IL measurements, using a silk suture on the antimesenteric border, were prospectively made on patients between 24 weeks of gestational age and 5 years of age undergoing laparotomy. Patients with gastrointestinal malformations or those above or below 2 SDs for growth parameters were excluded. A curve fitting process was applied to determine the best model for IL (small bowel and colon separately) from among postconception age, weight, and height at surgery. RESULTS One hundred eight patients participated in this study. Highly predictive (R(2) > 0.8) models for IL were determined for all predictor variables (postconception age, weight, and height) examined suggesting that all of these variables are excellent predictors determinants of IL. Although all models had statistically similar properties, the model using height had the best performance across the full range of the variable. CONCLUSION Although age, weight, nor height was definitely superior for the prediction of IL, we propose that until external validations of our models occur, height at surgery be used for the prediction of expected small intestinal and colon length in infants.

Ethanol lock therapy to reduce the incidence of catheter-related bloodstream infections in home parenteral nutrition patients with intestinal failure: preliminary experience☆

BACKGROUND Catheter-related bloodstream infections (CRBSI) cause morbidity and mortality in patients with intestinal failure dependent on parenteral nutrition. Ethanol lock of central venous catheters may decrease CRBSI, but limited pediatric data are available. METHODS Home parenteral nutrition patients with at least one previous CRBSI were initiated on a 70% ethanol lock protocol for a minimum of 4 hours. Infection rates (per 1000 catheter days) before and after initiation of ethanol locks were compared using a paired t test. RESULTS Ten patients (4 girls; median age, 44 months [range, 31-129 months]) began ethanol lock therapy after a total of 91 CRBSIs (37 gram-positive, 30 gram-negative, and 24 fungal) with a mean of 10.2 ± 6.2 per 1000 catheter days. Patients received ethanol lock for an average of 227 ± 64 days with only 3 CRBSI occurring (CRBSI rate of 0.9 ± 1.8 per 1000 catheter days [P = .005]). Central venous catheter replacements decreased from 5.6 per 1000 days to 0.3 per 1000 days posttherapy (P = .038). Ethanol lock was discontinued in 2 of 10 patients because of catheter thrombosis. CONCLUSION Preliminary results demonstrate a significant decrease in CRBSI with a 70% ethanol lock protocol. Catheter thrombosis may be a limitation that needs to be addressed. With such a dramatic therapeutic effect, a randomized trial is feasible and should be performed.

Intermediate outcomes after serial transverse enteroplasty in children with short bowel syndrome

AIM The aim of this study is to report 5-year outcomes in patients who have received serial transverse enteroplasty (STEP) using clinical and biochemical assessments of intestinal function. METHODS Thirty-one STEP procedures have been performed at our institution. Twelve patients who received STEP before January 2005 were analyzed. Clinical and biochemical outcomes were performed pre-STEP and post-STEP on an annual basis. Paired t tests were used to compare post-STEP outcomes to pre-STEP values (P < .05 was significant). MAIN RESULTS There were 12 patients (3 females; median age, 5.5 months). Serial transverse enteroplasty resulted in mean increase in length of dilated bowel segment of 89% ± 26% and increase in total small bowel length of 46% ± 40%. Two patients received liver-intestinal transplants post-STEP (at 4 and 5 months), and 2 patients died of liver failure at 3 and 8 months. The remaining 8 of 12 patients all show stable intestinal absorptive capacity (data not shown). Of 8 patients, 7 were weaned off parenteral nutrition by 4 years. No patient has required repeat STEP or bowel tapering. Two patients developed staple line leak, and 1 patient developed gastrointestinal bleed from staple line ulcers. CONCLUSION This report represents the largest series of STEP patients followed up beyond 5 years. Long-term clinical, biochemical, and radiologic outcomes appear to be robust.

The Role of Parenteral Lipids in the Development of Advanced Intestinal Failure–Associated Liver Disease in Infants A Multiple-Variable Analysis

BACKGROUND Given the recent interest in the role of ω-6 lipids in the development of intestinal failure-associated liver disease (IFALD), the authors sought to examine the role of parenteral lipids in the development of a serum conjugated bilirubin >100 µmol/L (5.9 mg/dL; CB100) in infants. METHOD Between 2003 and 2004, data were collected prospectively on infants undergoing an abdominal surgical procedure. Univariate logistic regression models for the prediction of CB100 by 1 year postoperatively were developed. Predictors significant at the 0.2 level on univariate analysis were entered into a backward stepwise multiple variable logistic regression. RESULTS Of 152 infants who received parenteral nutrition (PN) postoperatively, 22 developed CB100. Predictors that met criteria for consideration in the multiple-variable model were age, weight, small bowel length, presence of a stoma, proportion of enteral feeds postoperatively, septic episodes, days of maximal PN amino acid (>2.5 g/kg/d), days of maximal lipid (>2.5 g/kg/d), and PN duration. The final model included septic episodes (odds ratio, 3.23; 95% confidence interval, 1.8-5.9) and days of lipid >2.5 g/kg/d (1.04; 1.003-1.06). At 60 days of maximal lipid, the odds of advanced IFALD were increased 10-fold. CONCLUSIONS This model suggests a key role of parenteral lipids and septic events in the development of CB100 from IFALD. These data may provide targets, such as careful line care, reduction in maximal lipid dose, or alternate lipids such as ω-3 fatty acids, to prevent CB100, an identified marker of subsequent liver failure from IFALD.

Does the colon play a role in intestinal adaptation in infants with short bowel syndrome? A multiple variable analysis.

PURPOSE We sought to examine in a multiple variable model the impact of residual colonic length on time to intestinal adaptation in a cohort of infants with short bowel syndrome. METHOD Infants with a surgical diagnosis of short bowel syndrome who underwent operation 90 days or younger were included in this analysis. Univariate Cox proportional hazards models for time to full-enteral feeds were developed. Predictors significant at the .2 level were entered into a stepwise multiple variable Cox proportional hazards model. RESULTS A total of 106 infants were included in the cohort (70 adapted). Predictors meeting the criteria for the multiple variable model were as follows: multidisciplinary management (P = .045), Serial Transverse Enteroplasty Procedure (P = .057), percent small bowel (P < .001), percent large bowel (P < .001), preserved ileocecal valve (P = .001), number of septic (P < .001), and central line complications (P < .001). The final model included the following: multidisciplinary management (hazard ratio [HR], 1.932; 95% confidence interval [CI], 1.137-3.281), percent small bowel (HR, 1.028; 95% CI, 1.02-1.04), and septic events (HR, 0.695; 95% CI, 0.6-0.805). CONCLUSIONS The colon does not seem to play a significant role in intestinal adaptation. However, in addition to highlighting the importance of residual small bowel length, our model highlights the benefit of multidisciplinary intestinal rehabilitation and reduction of septic complications in achieving intestinal adaptation.

Human torovirus: a new virus associated with neonatal necrotizing enterocolitis.

AIM Toroviruses have been associated with gastroenteritis in both animals and humans. The aim of this study was to examine the fecal excretion of torovirus in infants with necrotizing enterocolitis (NEC). METHODS We reviewed all infants with NEC admitted to our tertiary care NICU over a 5-y period who had stool specimens sent for microbial culture and virology. Infants in the NICU during the same period with diagnoses other than NEC served as controls. RESULTS Forty-four infants with NEC stages I-III were identified, and pathogenic organisms were identified in 27 (61%). Toroviruses were identified in stool cultures in 48% of patients with NEC, and 17% of the non-NEC controls (p<0.001). There was no significant difference in illness severity or mortality between the torovirus-positive and -negative infants with NEC. CONCLUSION Torovirus should be added to the list of infectious agents associated with NEC in newborn infants. The exact role torovirus plays in the etiology and progression of NEC warrants further investigation.

Delayed primary serial transverse enteroplasty as a novel management strategy for infants with congenital ultra-short bowel syndrome.

BACKGROUND Congenital ultra-short bowel syndrome (USBS) is a challenging problem with a poor outcome. We report a new management approach for USBS infants that attempts to optimize gut growth potential. METHODS We report five neonates with USBS in whom no correction was performed at primary surgery except placement of a gastrostomy (G) tube. Sham feeds were started with intermittent G-tube clamping to induce bowel dilatation/growth. Serial fluoroscopy was done until bowel caliber reached 5 cm. STEP was performed and continuity established to the colonic remnant. Small bowel length (SBL) and enteral caloric intake were tabulated. RESULTS Patients were born with a mean residual SBL of 19 ± 7.6 cm (14.8% of expected). Median duration of sham feeds prior to STEP was 108 (range 27-232)days. Mean SBL at STEP was 47 ± 12.1cm, which increased post-STEP to 70 ± 12.7 cm (a mean increase of 296% from birth, representing 36.4% ± 13.1% of expected gut length). With a median follow-up time of 20 months (range 8-28), 4/5 achieved >50% enteral calories and have normal liver function. One has undergone liver transplantation. CONCLUSIONS In USBS patients, delayed surgical correction with sham feeds accelerates gut growth, optimizing potential for autologous reconstruction. This approach may offer greater opportunity for intestinal adaptation than traditional options.

Peripherally inserted central catheters for long-term parenteral nutrition in infants with intestinal failure.

Background and Aim: Infants with intestinal failure often require long-term central access for delivery of parenteral nutrition (PN). Traditionally, surgically placed central venous catheters (CVCs) have been used; however, the complications associated with these catheters can lead to significant morbidity. Peripherally inserted central catheters (PICCs) are potentially superior to CVCs because they tend to be smaller, and can be placed without general anesthesia. The purpose of the study is to report the use of PICCs for long-term administration of PN in infants with intestinal failure and compare with previously published catheter infection and venous thrombosis rates. Methods: A 4-year review of infants younger than 12 months with intestinal failure and a PICC for PN delivery was performed to determine the incidence of catheter-related bloodstream infections (CRBSIs) and PICC-associated venous thrombosis. The complication rates were compared with those reported for CVCs and PICCs in the pediatric literature. Results: A total of 45 infants with intestinal failure, receiving PN through a PICC were included in the study. Data from 95 PICCs accounting for 10,189 catheter days were collected. The overall incidence of CRBSI was 5.3/1000 catheter days and the incidence of venous thrombosis was 2.0/1000 catheter days. Conclusions: PICCs offer an advantage over CVCs in that they can often be inserted without a general anesthesia and do not require manipulation of the vein. Given the low rate of CRBSI and venous thrombosis, we recommend PICCs for infants with intestinal failure requiring PN.

Is the use of parenteral ω-3 lipid emulsions justified in surgical neonates with mild parenteral nutrition–associated liver dysfunction?

PURPOSE Although evidence suggests that parenteral omega-3 lipid emulsions (O-3LEs) may be beneficial in treating advanced parenteral nutrition (PN)-associated liver disease, our objective was to determine if O-3LEs are justified in those with early liver disease. METHODS This is a retrospective analysis of prospectively collected data on all surgical neonates, who received more than 1 day of PN postoperatively between 2001 and 2004 with observation through 2005 (era before O-3LE introduction). We examined the proportion of those who developed mild and advanced liver dysfunction. RESULTS Of the 292 infants in the cohort, 104 (36%) developed mild liver dysfunction (conjugated bilirubin, 34 micromol/L [cBili34]) after a mean of 22 days. Thirty-one (30%) of the cBili34 patients reached a serum conjugated bilirubin of 100 micromol/L, and 13 (13%) developed liver failure. Of these, 4 underwent transplantation, and 5 died of hepatic disease. Overall, 86 of the cBili34 patients (83%) were weaned off PN. CONCLUSION With more than 80% of cBili34 patients being weaned from PN without adverse hepatic sequelae, it is difficult, in the absence of definitive evidence of efficacy and safety for O-3LEs together with increased costs, to justify the routine use of O-3LEs in this low-risk population outside formal research protocols.

Early neurodevelopmental outcomes of infants with intestinal failure.

BACKGROUND The survival rate of infants and children with intestinal failure is increasing, necessitating a greater focus on their developmental trajectory. AIMS To evaluate neurodevelopmental outcomes in children with intestinal failure at 0-15months corrected age. STUDY DESIGN Analysis of clinical, demographic and developmental assessment results of 33 children followed in an intestinal rehabilitation program between 2011 and 2014. Outcome measures included: Prechtls Assessment of General Movements, Movement Assessment of Infants, Alberta Infant Motor Scale and Mullen Scales of Early Learning. Clinical factors were correlated with poorer developmental outcomes at 12-15months corrected age. RESULTS Thirty-three infants (17 males), median gestational age 34weeks (interquartile range 29.5-36.0) with birth weight 1.98kg (interquartile range 1.17-2.50). Twenty-nine (88%) infants had abnormal General Movements. More than half had suspect or abnormal scores on the Alberta Infant Motor Scale and medium to high-risk scores for future neuromotor delay on the Movement Assessment of Infants. Delays were seen across all Mullen subscales, most notably in gross motor skills. Factors significantly associated with poorer outcomes at 12-15months included: prematurity, low birth weight, central nervous system co-morbidity, longer neonatal intensive care admission, necrotizing enterocolitis diagnosis, number of operations and conjugated hyperbilirubinemia. CONCLUSION Multiple risk factors contribute to early developmental delay in children with intestinal failure, highlighting the importance of close developmental follow-up.